 |
Because neurologists felt that surgeons performed too many inappropriate operations, angiographic/minor procedural strokes were excluded from quoted risks and surgeons cited the worst natural history outcomes to justify practice! By contrast, many surgeons felt RCTs to be unnecessary and that they challenged the principle of ‘intuitive reasoning’ that had served them so well for so long. Anything sound familiar regarding current debates? |
Unfortunately, due to a variety of reasons including trials stopping prematurely, conflicts of interest and possible interventionist inexperience, eight RCTs comparing CEA with carotid artery stenting (CAS) failed to establish consensus.
Disappointing...
It is disappointing that people still believe registries can replace RCTs. For the record,
- no angioplasty RCT/Registry has reported an angiographic stroke
- none of the huge ELOCAS, ALKK, Pro-CAS and Global Stent Registries report 30-day all death/any stroke
- the 30-day complication rates in ‘high risk for CEA’ asymptomatic patients in SAPPHIRE and ARCHeR were sufficiently high (>5%) that no long term reduction in stroke could accrue
- CAPTURE and ARCHeR reported >10% complication rates in symptomatic patients (again, questionable long term benefit)
- while the latest National Inpatient Sample reported a startling in-hospital 7.5% mortality in >1,700 symptomatic patients undergoing CAS in the United States.
CAS has great potential for treating truly high risk patients, but it cannot be justified if high procedural risks cancel out any prospect of long term benefit
There is no evidence that standard risk, symptomatic/asymptomatic patients should be treated by CAS outwith RCTs. CAS is, however, likely to become the preferred intervention in high risk symptomatic patients, but the 30-day risk of stroke/death after CAS has to be kept <10% for long term prevention of stroke to be achieved. Similarly, unless the 30-day death/stroke is kept <4% in asymptomatic ‘high risk for CEA’ patients undergoing CAS, no long term prevention of stroke will occur (i.e. it may be better to treat these patients conservatively).
In short, CAS has great potential for treating truly high risk patients, but it cannot be justified if high procedural risks cancel out any prospect of long term benefit. This simple fact has always been the Achilles’ heel of carotid endarterectomy.
It is inevitable that CEA and CAS will evolve complementary roles, but we are deluding ourselves if we think we already know who benefits most from CAS. Just look again at those recent National Inpatient Sample mortality rates.
Please support CREST and ICSS (RCTs in symptomatic patients) and ACT-1, TACIT, ACST II and CREST (RCTs in asymptomatic patients). Registries and ‘intuitive reasoning’ are no substitute for large, properly conducted RCTs.